Thirty years and 1663 consecutive Norwood procedures: Has survival plateaued?

Hypoplastic left heart syndrome is one of the most common and challenging lesions requiring surgical intervention in the neonatal period. The Norwood procedure for hypoplastic left heart syndrome was first reported in 1983. The objective of this study was to describe early outcomes after the Norwood procedure at a single institution over 30 years.

This retrospective cohort study included all patients with hypoplastic left heart syndrome (and variants) who underwent the Norwood procedure between January 1984 and May 2014 at a single institution. The study period was divided into 6 eras: era 1, 1984 to 1988; era 2, 1989 to 1993; era 3, 1994 to 1998; era 4, 1999 to 2003; era 5, 2004 to 2008; and era 6, 2009 to 2014. The primary outcome was in-hospital mortality after the Norwood procedure. Binomial point estimates complete with 95% confidence intervals (CL_0.95) were computed for the entire cohort and by era.

During the study period, 1663 infants underwent the Norwood procedure. Overall in-hospital mortality was 25.9% (CL_0.95, 23.8-28.0). Mortality by chronologic era was 40.4% (CL_0.95, 34.9-45.9), 33.6% (CL_0.95, 29.2-37.9), 28.7% (CL_0.95, 22.8-34.6), 14.9% (CL_0.95, 10.4-19.3), 11.2% (CL_0.95, 7.4-15.0), and 15.7% (CL_0.95, 10.3-21.1). Survival was improved in eras 4 to 6 compared with eras 1 to 3 (P all < .03). Anomalous pulmonary drainage, moderate to severe atrioventricular valve regurgitation, lower birth weight, earlier era, younger gestational age, genetic anomaly, preterm birth, race other than white or African-American, and lower weight at the Norwood procedure were associated with increased mortality. Mortality was greatest in patients with 3 or more risk factors. In the best-fitting multiple covariate model, anomalous pulmonary venous drainage, gestational age in weeks, genetic anomaly, and race other than white and African American were statistically significant contributors, after adjusting for era.

Survival after the Norwood procedure has plateaued despite improvements in diagnosis, perioperative care, and surgical techniques. Nonmodifiable patient characteristics are important determinants of the risk of mortality.